Treffer: Audio processing and sound synthesis in Haskell

Our goal is to develop a purely functional audio processing and sound synthesis framework using Haskell, that is more expressive than any existing framework, and that is at least as efficient. Expressiveness and performance are of equal importance to us. Indeed, with modern-day processors, we believe that real-time performance can be achieved. A decade or two ago, the computational demand for real-time sound synthesis in software exceeded the capability of conventional processors. Software sound synthesis was usually done by using dedicated audio processing languages such as csound [Ver91] or written in low-level languages such as C. Sophisticated audio synthesis in real-time was often not feasible, and was usually done in hardware. Over the years, with the advent of ever faster processors and state-of-the-art compiler technologies for functional languages, real-time audio synthesis using a purely functional language has gradually become within reach. A common source of inefficiency in functional programs is unexpected memory usage, otherwise known as space leaks. A space leak in an audio processing